Essays on Market Liquidity and Time-Varying Jump Dynamics in the Stock Market

Market Liquidity and Time-Varying Jump Intensity Dynamics in Aggregate Stock Market Returns: I find that significant time variations in the aggregate stock market's jump intensity are partly attributable to a market liquidity dynamic, although a latent information dynamic captured by an approximate autoregressive moving average (ARMA) form of stock market returns is also driving the jump intensity process. In magnitude, the relative significance of market liquidity is 11.7% of jump intensity on average but reaches 40% for some time periods. I also find that the model that features both a market liquidity dynamic and an approximate ARMA form of stock returns in its jump intensity process more accurately estimate the underlying volatility dynamic of the aggregate stock market than other models. My research highlights the importance of market frictions in modeling the aggregate volatility process.

Aggregate Stock Market's Jump Intensity in the Cross-Section of Expected Stock Returns: Consistent with the extant economic theory, I find that stocks with high sensitivities to innovations in the aggregate stock market's jump intensity have low average returns, and size, book-to-market, and aggregate volatility cannot account for the low average returns earned by these stocks during the period from October 1987 to June 2009. I also find that a factor that mimics these innovations carries an economically and statistically significant negative price of risk. In magnitude, a two-standard deviation increase in the jump loadings of a diversified stock portfolio leads to a 3.9% decrease in its expected annualized returns. Such relations do not hold if the jump dynamic is driven by a market liquidity dynamic. They are also time-varying and do not hold during the period either before the Black Monday crash in October 1987 or after the end of the most recent financial crisis in June 2009.